Sexually transmitted Chlamydia trachomatis infection is extremely common and commonly produces salpingitis in infected women. Seroepidemiologic data suggest that post-salpingitis sequela of C.trachomatis infection such as tubal infertility or ectopic pregnancy are accompanied by immune responses to the chlamydial heat shock protein 60 (hsp60) antigen. Animal experiments suggest that chlamydial disease is attributable to immunopathologic responses to the chlamydial hsp60 antigen. Immune responses to hsp60 are genetically determined and in the mouse linked in part to the H-2 locus. Based on these observations, the hypothesis of the present proposal is that HLA genes, by determining antibody responses to the chlamydial hsp60 in humans, are a risk factor for clinical salpingitis following chlamydial cervical infection. HIV infection is speculated to interrupt this pathogenetic pathway by reducing immune responses to hsp60. To address this hypothesis, 300 women (approximately 200 HIV infected and 100 HIV uninfected) who are commercial sex workers in the Pumwani district of Nairobi, Kenya will be followed for up to four years for incident infections with c.trachomatis and for the occurrence of clinical salpingitis. All women will be HLA typed (class I by standard serologic tests and class II by DR, DP and DQ oligotyping) and have baseline serologic evaluation for antibodies to chlamydial hsp60. Cloned recombinant chlamydial hsp60 and synthetic peptides will be used in a variety of serologic formats. The sample size is sufficient to detect a increased relative risk of greater than or equal to 1.86 if hsp60 antibodies are correlated with a specific HLA allele and to detect an increased relative risk of greater than or equal to 1.68 if acute salpingitis is associated with an antibody response to the chlamydial hsp60. This study should define the molecular and genetic basis for chlamydial salpingitis and may help define new approaches to improved treatment or prevention of this disease.